AI Frameworks Engineer

TL;DR

AI Frameworks Engineer: Developing and optimizing deep learning compiler technology for hirify.global NPUs with an accent on MLIR-based compilers and AI execution middleware. Focus on performance tuning, developing optimization techniques for neural networks, and ensuring high-quality, high-performance product software.

Location: Hybrid in Leixlip, Ireland

Company

hirify.global is dedicated to developing hardware accelerators for the future of AI, enabling machine hirify.globalligence on edge, desktop, and mobile computers.

What you will do

• Develop MLIR-based compiler technology for deep learning workloads on hirify.global NPUs.
• Develop AI execution middleware based on hirify.global OpenVINO toolkit.
• Develop large-scale production software with validation and continuous integration.
• Collaborate with deep learning software framework teams and hardware architecture teams to accelerate next-generation software.
• Coordinate internally and externally with cross-geographical teams such as execution runtime software, NPU hardware, infrastructure, and front-end teams.

Requirements

• BS/MS in Computer Science or a similar field.
• At least 4 years of experience in C++ programming, software design, debugging, performance analysis, and test design.
• Excellent data structure, algorithms, and modern C++ programming skills.
• Experience in one of the following: compiler technologies, deep-learning frameworks or algorithms, high-performance computing, computer vision, or numerical modeling.
• Positive can-do attitude, desire to deliver results, and passion for optimization and performance.
• Excellent written and oral English communication skills.

Nice to have

• Experience in developing modern compilers, especially LLVM and MLIR.
• Experience with formulating optimization problems and using ILP solvers (e.g., ORTOOLs, CPLEX, SCIP).
• Experience in AI hardware accelerators, GPU, or heterogeneous architectures software development.
• Background in performance analysis and optimization, particularly in machine learning.
• Solid understanding of modern machine-learning primitives and LLMs.
• Background in Python, modern AI frameworks, and ecosystem (e.g., Torch, HuggingFace, llama.cpp).

Culture & Benefits

• Immerse yourself in a diverse, encouraging environment.
• Join the adventure of harnessing state-of-the-art deep neural networks and advanced AI hardware.
• Work on building the world's leading deep learning compiler.
• Benefit from a hybrid work model allowing split time between on-site at hirify.global Leixlip and off-site.